| Abstract |
Traditionally, probability integral transforms (PITs) have been popular means for evaluating density forecasts. For an ideal density forecast, the PITs should be uniformly distributed on the unit interval and independent. However, this is only a necessary condition, and not a sufficient one, as shown by some simple examples. I discuss an alternative approach to density forecast evaluation, via the Kullback-Leibler information criterion (KLIC), and illustrate it with a small simulation study.
|
